ABSTRACT
The chapter begins by justifying the adopted approach to reference against a traditional and still widely pursued conception of it. According to the approach to reference taken in Chapter 3, it models linguistic understanding; the alternative conception has it that the reference of an expression is what relevant to determining truth-values. It is argued that this latter approach will not work. There follows a discussion of the way in which the notions of sense and reference apply to the sentence and to its semantically significant parts. Frege was right to identify the referents of predicates with concepts, but he should have taken the referents of declarative sentences to be Russellian propositions rather than truth-values. It is considered to what extent referents of predicates and sentences can be identified in turn with so-called Carnapian intensions. There is then a discussion of the difference between the semantics of genuine proper names and definite descriptions, taking in the notions of rigidity and the de re. It is argued that the notion of the de re is fundamental to any satisfactory philosophical account of the way in which it is possible for thought to target a world, but that in the context of the favoured conception of reference it is a technical superfluity, since all reference is de re. The differences between genuine names and descriptive names is examined in detail, and the chapter concludes with a brief consideration of the role of the acquaintance relation in singular thought.
